| 01:09:36 | arkygeek: | hi everyone |
| 01:10:05 | arkygeek: | basic question .... ---> How do I add a primary key to a table? |
| 01:10:47 | arkygeek: | I have a column i want to use, but it is of type real at the moment. |
| 03:39:13 | grobda24: | Is there a way of getting text from polygon type without doing this .... http://etutorials.org/SQL/Mastering+Oracle+SQL/Chapter+17.+Regular+Expressions/17.3+Advanced+Regular+Expression+Syntax/ ...( pulls hair out ). |
| 03:39:15 | sigq: | Title: 17.3 Advanced Regular Expression Syntax :: Chapter 17. Regular Expressions :: Mastering Oracle SQL :: SQL :: eTutorials.org ( at etutorials.org ) |
| 03:44:19 | mloskot: | grobda24: text of polygon type? |
| 03:44:32 | mloskot: | You mean name of geometry type or geometry content in WKT form ? |
| 03:53:23 | grobda24: | mloskot, hello. WKT = text ? If so then yes. The content of a Polygon type as text. Syntax: polygonfromtext( 'POLYGON( ( 0 0, 1 0, 1 1, 0 0 ) )', -1 ); ... of course the point numbers would come from a *POSTGRESQL* polygon column. |
| 03:55:21 | mloskot: | grobda24: you know what is WKT aka Well-Known-Text ? |
| 03:56:39 | grobda24: | mlostkot ... I think so. Basic text ? I'm learning here as well as I'm from a MySQL background. |
| 03:57:36 | mloskot: | grobda24: WKT is defined by OpenGIS Simple Feature Specification, so it's not specific to any DBMS |
| 03:58:01 | grobda24: | k |
| 03:58:09 | mloskot: | If you are not sure about WKT, my suggestion is to learn about it first, otherwise you are likely going to struggle with using spatial stuff in postgis/mysql |
| 03:58:13 | mloskot: | http://www.postgis.org/documentation/manual-1.3/ch04.html#id2978551 |
| 03:58:14 | sigq: | Title: Chapter 4. Using PostGIS ( at www.postgis.org ) |
| 03:58:29 | mloskot: | http://www.opengeospatial.org/standards/sfs |
| 03:58:30 | sigq: | Title: Simple Feature Access - Part 2: SQL Option | OGCĀ® ( at www.opengeospatial.org ) |
| 03:59:03 | grobda24: | I'm just going to import my source as text. It doesn't look as easy to reference a polygon. A point can be referenced as an array. |
| 03:59:50 | mloskot: | 1. understand what WKT is |
| 04:00:09 | mloskot: | 2. read postgis manual and its functions reference |
| 04:00:16 | mloskot: | 3. learn about ST_AsText to get textual representation of geometry |
| 04:00:38 | mloskot: | http://www.postgis.org/documentation/manual-1.3/ch06.html |
| 04:00:39 | sigq: | Title: Chapter 6. PostGIS Reference ( at www.postgis.org ) |
| 04:01:02 | grobda24: | k |
| 04:13:24 | grobda24: | No, no. I don't think mlostkot got what I was saying. I'm wanted to get text from a standard PostgreSQL polygon column NOT a PostGIS one. |
| 04:13:46 | grobda24: | Anyway, I have another solution : ) |
| 09:26:19 | darkblue_B: | well, for whatever reason.. I just changed the test files at svg.cc/pg/occ so they actually display correctly in my browsers |
| 09:26:28 | darkblue_B: | my set of those files I should say |
| 09:27:10 | darkblue_B: | just about everything was wrong with them.. as far as I could tell, I tried a few browsers and none of them showed the svg correctly |
| 09:27:19 | darkblue_B: | now it works |
| 09:29:51 | darkblue_B: | hi mloskot |
| 09:30:07 | mloskot: | hi |
| 09:31:27 | darkblue_B: | do you have a way you execute PostGIS queries and see the results visually? for example, the OGC SFS Test suite, in Postgis exmaples |
| 09:31:52 | mloskot: | not really |
| 09:32:02 | mloskot: | JTS validation suite may be helpful |
| 09:32:09 | mloskot: | to understand geometry engine |
| 09:32:10 | darkblue_B: | yeah.. ok |
| 09:32:18 | mloskot: | it's available on the web |
| 09:32:41 | darkblue_B: | I found a set of generated svg at svg.cc/pg/occ |
| 09:32:59 | darkblue_B: | just wondering about tools and visualizations |
| 09:33:17 | darkblue_B: | thx |
| 09:34:07 | mloskot: | http://www.vividsolutions.com/jts/tests/index.html |
| 09:34:09 | sigq: | Title: JTS Validation Suite ( at www.vividsolutions.com ) |
| 09:36:01 | darkblue_B: | hmm vividsolutions is not responding for me |
| 09:36:12 | darkblue_B: | but no matter... not important right now |
| 11:12:10 | sigq: | geosfeed: Ticket #187 ( defect updated ): ST_Buffer crashes on WinXP, PostgreSQL 8.3, PostGIS 1.3.3, GEOS 3.0.0-CAPI-1.4.1 <http://trac.osgeo.org/geos/ticket/187#comment:2> |
| 12:04:20 | johto: | hello. how could I "substract" a geometry from another geometry? |
| 12:06:27 | mloskot: | johto: give some example |
| 12:14:34 | johto: | mloskot; for example, I had a smaller rectangle inside a bigger rectangle and only wanted the part around the smaller rectangle |
| 12:15:30 | mloskot: | johto: check this |
| 12:15:31 | mloskot: | http://www.postgis.org/documentation/manual-1.3/ch06.html#id2982990 |
| 12:15:32 | sigq: | Title: Chapter 6. PostGIS Reference ( at www.postgis.org ) |
| 12:15:55 | mloskot: | I suppose you will want to play with functions like ST_InteriorRingN, etc. |
| 12:16:09 | mloskot: | or ST_GeometryN |
| 12:16:15 | mloskot: | and others from this category |
| 15:11:30 | johto: | getting number of vertices in a multipolygon? |
| 15:13:00 | tazle: | sum( st_numpoints( st_dump( mp_geom_column ).geom ) ) or something |
| 15:13:16 | tazle: | you'd also need group by, I think |
| 15:13:54 | tazle: | ie. there is no direct way - explode the multipolygon into polygons, count the points of those and sum |
| 15:14:34 | darkblue_B: | cheating completely, I read the polys AsText( ) and counted commas and parens |
| 15:15:12 | johto: | yeah but I'd need them in an ORDER BY clause |
| 15:15:14 | darkblue_B: | using a scripting language.. |
| 15:15:50 | johto: | and transferring millions of points isn't something i want to consider |
| 15:16:21 | darkblue_B: | millions? |
| 15:16:37 | johto: | well, I think it's closer to billions |
| 15:17:05 | darkblue_B: | I was asked to run some checkng on a data set.. |
| 15:17:20 | darkblue_B: | so I wrote in python.. did the Valid( ) and Simple( ) queries |
| 15:17:41 | darkblue_B: | as above, AsText( ).. and then did my analysis in python |
| 15:18:13 | johto: | maybe I should share my scenario.. |
| 15:18:15 | darkblue_B: | if you really have billions them write back to a new table |
| 15:18:28 | tazle: | johto: please do |
| 15:18:55 | tazle: | johto: also, is globepointer.com in Finland and doing something interesting?- ) |
| 15:19:13 | johto: | I have a polygon which describes the whole land area, and a number of polygons which describe the water area |
| 15:19:19 | johto: | tazle; yes, and yes |
| 15:19:29 | tazle: | johto: the material from MML? |
| 15:19:33 | johto: | mml? |
| 15:19:35 | darkblue_B: | ooh I remember this data set from last Fall |
| 15:19:37 | tazle: | Maanmittauslaitos |
| 15:19:42 | johto: | tazle; no, tele atlas |
| 15:20:27 | johto: | so I have to remove the water parts from the land. now I'm just calculating the ST_Difference( landpoly, waterpoly ) for every water polygon |
| 15:20:31 | darkblue_B: | so the land poly really is some huge number of vertices |
| 15:20:40 | johto: | no, the water polys are |
| 15:20:47 | tazle: | johto: perhaps I should ask "why?" |
| 15:21:00 | johto: | tazle; for rendering. |
| 15:21:17 | tazle: | johto: render the land polygon first, then the water polygons on it |
| 15:21:28 | darkblue_B: | rasterize it sooner rather than later |
| 15:21:31 | johto: | if I first render the land polygon and then the water polygons on top of it, antialiasing gives me ugly outer rings |
| 15:21:47 | tazle: | johto: example? |
| 15:21:53 | johto: | tazle; sec |
| 15:22:48 | darkblue_B: | didnt we talk about slicing the data set.... |
| 15:23:06 | darkblue_B: | into subsets... |
| 15:23:33 | johto: | http://www.neonaspect.com/~rnn/ugly.png |
| 15:24:02 | tazle: | johto: oah yes - cobmine your water polygons |
| 15:24:43 | tazle: | johto: also, are you supposed to render in real-time, or just once into tiles? |
| 15:25:05 | johto: | prerender quarterly into tiles |
| 15:25:27 | johto: | ( i.e. every time tele atlas provides us with new data ) |
| 15:25:41 | tazle: | ok, then just combining the water polygons is likely the quickest solution |
| 15:26:08 | johto: | quickest processing or rendering? |
| 15:26:17 | tazle: | both, most likely |
| 15:26:38 | tazle: | a land polygon with all the lakes of Finland carved out of it would be huge |
| 15:28:13 | johto: | well, that way we wouldn't have to render all the land and water on top of each other |
| 15:28:20 | johto: | ( the way I'm thinking right now ) |
| 15:28:42 | tazle: | also, just combining the water areas produces huge polygons, I have some ideas, but since you are a competitor, I'm somewhat wary of sharing them |
| 15:28:43 | johto: | and it wouldn't have to be one huge polygon, all the islands for example could be taken out |
| 15:28:55 | johto: | tazle; ; ) |
| 15:29:03 | johto: | tazle; what company are you working for? |
| 15:29:07 | tazle: | johto: Indagon |
| 15:29:11 | johto: | oh |
| 15:29:22 | tazle: | yeah : ) |
| 15:30:00 | johto: | do you use google maps btw.? |
| 15:30:30 | tazle: | not us, the map on the front page is from a partner |
| 15:31:04 | johto: | what did you use for rendering the tiles? |
| 15:31:06 | johto: | or do you |
| 15:31:09 | johto: | *do use |
| 15:31:34 | tazle: | no, we ship vector maps |
| 15:31:57 | johto: | oh. so you don't have any browser interface? |
| 15:33:08 | tazle: | not a general browser interface, no |
| 15:36:15 | tazle: | if I were doing a web interface, I would look into Google Maps, though |
| 15:36:42 | tazle: | for a limited user base, the terms are not *so* bad |
| 15:37:39 | johto: | yeah. but it's too late now |
| 15:38:04 | tazle: | yeah, if you've already shelled out the cash for TeleAtlas material, I woul dsay you're stuck with them for a while at least : ) |
| 15:46:31 | johto: | btw. there is an ST_npoints( ) function |
| 15:46:38 | johto: | which seems to return the number of vertices |
| 16:11:50 | tazle: | so it would seem, a postgis extension |
| 19:04:17 | TheSteveMonkey: | I want to check my thinking on a query - I have a zoning tables and a parcel table and if I want to find all the parcels within a certain zone - do I need to union the zone geometries first? |
| 19:04:31 | TheSteveMonkey: | It seems like most of the spatial join functions do not accept geometry collections |
| 19:04:41 | TheSteveMonkey: | here is what I am using but it is incredibly slow |
| 19:04:53 | TheSteveMonkey: | select p.gid from parcels as p, zoning as z where ST_Within( ( select ST_Union( z.the_geom ) from zoning as z where z.comb_zn like 'A%' ), p.the_geom ); |
| 19:24:46 | TheSteveMonkey: | anyone? |
| 21:28:10 | jcooke: | Could anyone recommend a good tutorial/eBook/paper book for getting started with PostGIS or GIS in general, for someone with no kind of GIS experience at all. |
| 21:28:35 | TheSteveMonkey: | http://www.bostongis.com/?content_name=postgis_tut01#20 |
| 21:28:38 | sigq: | Title: Part 1: Getting Started With PostGIS: An almost Idiot's Guide ( at www.bostongis.com ) |
| 21:28:49 | TheSteveMonkey: | jcooke: where are you located? |
| 21:28:59 | jcooke: | Austin |
| 21:29:16 | TheSteveMonkey: | I am giving an intro preso at LinuxWorld SF on tuesday next week |
| 21:29:28 | TheSteveMonkey: | thought if you were close you might stop by - it's free |
| 21:29:41 | jcooke: | TheSteveMonkey, I've looked at that before, but it seems to assume some level of experience with GIS data. |
| 21:30:04 | jcooke: | SF is a bit out of the way : ) but thank you. |
| 21:30:54 | jcooke: | If that article is the best to start with though, that's fine, I'll just have to do a bit of research on some of the jargon, etc. |
| 21:34:42 | TheSteveMonkey: | You could do one of the Free ESRI intro to GIS tutorials - there is some ESRI lingo in their but there is also some general GIS Concepts |
| 21:35:31 | jcooke: | TheSteveMonkey, sweet, thanks! |
| 23:27:09 | darkblue_B: | hmmm I just used an AND in the WHERE |
| 23:30:27 | darkblue_B: | z.comb_zn like 'A%' and ST_Within( p.the_geom, z.the_geom ) |
| 23:30:42 | darkblue_B: | seems to work fine |
